Recent developments in dog (Canis familiaris) genomics have catapulted this species to the status of a model organism, with major advantages for the study of complex genetic diseases and traits relevant to the human condition. Its rapid rise in genomics was fueled by a 7.6X genome sequence, the identification of over 2 million single nucleotide polymorphisms (SNPs), and a commercial SNP microarray with over 125,000 loci. These developments converge with the enormous phenotypic variation between different breeds, the recent establishment of many breeds by line breeding that has yielded an advantageous haplotype structure for genetic analyses, greater physiologic and anatomic relationship to human diseases than other animal models, and excellent pedigree records. The main barrier to fully exploit this model for genetic studies is ready access to sufficient numbers of control and affected animals relevant to the traits of interest. The Cornell University Hospital for Animals (CUHA) admits - 15,000 canine patients (5,000 new visits) each year. We propose to create a DNA repository from selected pure-breed dogs admitted to our hospital accurately diagnosed with complex diseases and traits of strong biomedical interest. A conservative estimate is that approximately 2,000 DNA samples per year will be added to our existing collection, and a subset of up to 400 cases and controls per year will be subjected to high-density SNP genotyping. Initially, we will map loci for owner-directed aggression, rod-cone dystrophy, hepatic microvascular dysplasia, and granulomatous (ulcerative) colitis in specific breeds of dog. We will prospectively accumulate phenotypic data for the same specific breed controls. A standard phenotype screen will be performed on dogs over 8 years of age belonging to the same breeds admitted with the 12 most frequently diagnosed diseases in our hospital. We will encourage the canine genetic mapping community to submit requests for SNP genotyping of their cases and controls, on a cost-sharing basis. The Cornell Medical Genetic Archive will purchase an equal number of mapping arrays as supported by this grant application. Future mapping studies will concentrate on complex diseases of dogs among those breeds most frequently admitted to our hospital under the advice of highly qualified, external consultants. The SNP genotypes and accompanying phenotypes will be uploaded to the University of California Santa Cruz (UCSC) Genome Bioinformatics mirror and ported to the original UCSC site. We will help alleviate the need for scientists to repetitively collect control samples and we will provide SNP genotypes to allow in silico analyses at minimum cost. By combining a large collection of well phenotyped samples with SNP genotypes, this resource will facilitate the genetic analysis of complex traits, make the canine model accessible to a wide cadre of scientists, and empower the dog as an indispensable element in biomedical research.